Anesthesia type is a significant predictor of postoperative delirium in patients undergoing unilateral total knee arthroplasty.
According to a retrospective analysis of anesthesia billing data, International Classification of Diseases, Ninth Revision (ICD-9) codes and demographic information, patients who have received general anesthesia may be twice as likely to develop delirium postoperatively when compared with combined spinal-epidural, epidural or spinal patients. Given that other risk factors for postoperative delirium are difficult or impossible to alter, investigators underscored the importance of considering anesthesia type as a modifiable risk factor for postoperative delirium.
“ In this cohort of total knee arthroplasty patients, general anesthesia was associated with up to two times higher odds of delirium,” said Sarah M. Weinstein, a research data analyst at the Hospital for Special Surgery, in New York City. “Given that other risk factors might not be modifiable, anesthesia type is especially important to consider as a way to reduce risk for delirium as well as other complications.”
Modifiable Risk Factors
As Ms. Weinstein reported, postoperative delirium has drawn increased attention, with recent publications highlighting an association with longer hospital stays, higher cost of care and potentially increased mortality rates ( Gen Hosp Psychiatry 2015;37:223-229).
“Previous research has suggested that older age and psychiatric comorbidities increase risk for delirium, but the impact of modifiable risk factors, such as anesthesia type, is still debated,” Ms. Weinstein said. “We were thus interested in looking at the role of modifiable risk factors, particularly anesthesia type and risk for delirium. We hypothesized that patients who had general anesthesia would have higher odds of developing postoperative delirium than those who received neuraxial anesthesia.”
Following institutional review board approval, Ms. Weinstein and her colleagues identified patients who underwent unilateral total knee arthroplasty at the Hospital for Special Surgery between 2005 and 2014. Using statistical software, the researchers merged various administrative data sets, combining anesthesia billing data, ICD-9 codes from surgical billing and demographic information. The resulting data set consisted of 22,179 patients. The researchers excluded patients who didn’t have their surgery on the date they were admitted (n=73) and those whose primary type of anesthesia was missing from the records (n=901).
For the remaining 21,205 patients, Ms. Weinstein and her colleagues used multivariable logistic regression analysis to find adjusted odds ratios to predict postoperative delirium. Covariates included age, sex, type of anesthesia (general, combined spinal-epidural, epidural or spinal) and comorbidities: hypothyroidism, renal, diabetes, obesity, sleep apnea, valvular disease, hypertension, congestive heart failure, tobacco abuse, pulmonary, depression/anxiety and psychoses.
As Ms. Weinstein reported at the 2017 annual Spring meeting of the American Society of Regional Anesthesia and Pain Medicine (abstract 3585), 454 (2.14%) of the 21,205 patients in this cohort had an ICD-9 code indicating postoperative delirium diagnosis. Consistent with previous literature, patients with postoperative delirium were, on average, older and had a higher comorbidity burden than those without postoperative delirium.
After controlling for age, sex and comorbidities, type of anesthesia was found to be a significant predictor of delirium. Compared with patients who received general anesthesia, the adjusted odds ratios for combined spinal-epidural patients, epidural patients and spinal patients were 0.55 (95% CI, 0.34-0.89; P=0.0155), 0.45 (95% CI, 0.20-0.99; P=0.0483) and 0.52 (95% CI, 0.30-0.91; P=0.0211), respectively.
These data suggest that a patient’s odds of developing postoperative delirium were up to twice as high for general anesthesia patients compared with those who received neuraxial anesthesia, noted Ms. Weinstein, who added that further research is needed to validate these results and demonstrate the extent to which these findings are generalizable to other patient cohorts.
“These results are a portion of an ongoing study,” Ms. Weinstein said. “We’re interested in expanding this cohort to include other types of orthopedic patients. We’re also working on adding covariates from pharmacologic and laboratory data sets.”
David Birnbach, MD, MPH, vice provost and professor of anesthesiology, obstetrics and gynecology, and public health at the University of Miami Miller School of Medicine, asked whether the use of ketamine infusions was associated with an increased risk for delirium.
“We are looking at this question in a separate data set,” Ms. Weinstein said. “In previous analyses, though, we have found that ketamine was associated with increased risk for delirium. It’s only a small number of patients, but there was an association.”
“How do you control for changing sedation and multimodal techniques over time?” Dr. Birnbach asked. “Delirium outcomes may be influenced by use of less opioids, for example.”
“We try to adjust for these things by taking different time periods into account,” said Stavros G. Memtsoudis, MD, PhD, attending anesthesiologist and director of critical care at the Hospital for Special Surgery. “But there is ongoing work to tease out the use of multimodal, nonopioid analgesics and how that changes over time. So far, our conclusions haven’t changed when we add these modifiers.”